Research
Benchmarks
Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.
309,131 benchmarksPage 1460 of 12,366
Benchmarks
| Task | Dataset | Domain | Trend | Results | Last update |
|---|---|---|---|---|---|
| Super Resolution | ImageNet-256 (test) | 21 | Apr 21, 2026 | ||
| Agent Task | AppWorld Challenge (test) | 21 | Jun 16, 2026 | ||
| Multi-talker Automatic Speech Recognition | Libri2Mix clean (test) | 21 | Jul 8, 2026 | ||
| Color Denoising | DIV2K | 21 | Mar 30, 2026 | ||
| Survival Prediction | TCGA-BRCA (N=899) | 21 | Mar 30, 2026 | ||
| Survival Prediction | TCGA-BLCA N=347 | 21 | Mar 30, 2026 | ||
| Reasoning | Sudoku Extreme | 21 | May 21, 2026 | ||
| Open Vocabulary Mobile Pick and Place | OVMM Home Robot (val) | 21 | Jul 8, 2026 | ||
| Image-Text Retrieval | IIW (test) | 21 | Mar 27, 2026 | ||
| Compositional Evaluation | SugarCrepe | 21 | Apr 14, 2026 | ||
| Machine Translation | WMT News Commentary and IWSLT (En-De) 25 | 21 | Mar 27, 2026 | ||
| Multi-channel speech enhancement | Simulated adhoc microphone array dataset 4-channel (test) | 21 | Mar 27, 2026 | ||
| 3D Point Cloud Classification | MN40 | 21 | Mar 27, 2026 | ||
| Fact Verification | InfoTabs (held-out) | 21 | Mar 27, 2026 | ||
| Fact Verification | TabFact (held-in) | 21 | Mar 27, 2026 | ||
| Question Answering | TAT-QA (held-out) | 21 | Mar 27, 2026 | ||
| Question Answering | HiTab (held-in) | 21 | Mar 27, 2026 | ||
| Question Answering | WTQ (held-in) | 21 | Mar 27, 2026 | ||
| Question Answering | TabMWP (held-in) | 21 | Mar 27, 2026 | ||
| Audio Understanding | X-Ares | 21 | Mar 27, 2026 | ||
| Mathematical Reasoning | MathVision | 21 | Mar 27, 2026 | ||
| Chart Question Answering | ChartQA | 21 | Mar 27, 2026 | ||
| Science Question Answering | ScienceQA | 21 | Mar 27, 2026 | ||
| Outside Knowledge Visual Question Answering | AOKVQA | 21 | Mar 27, 2026 | ||
| Multimodal Understanding | MMMU | 21 | Mar 27, 2026 |